Aug 2 4 1987
Total Page:16
File Type:pdf, Size:1020Kb
(3 UCRL- 96157 .I PREPRINT ,II I, L r' AUG 2 4 1987 High Resolution Seismic Attenuation Tomography at Medicine Lake Volcano, California John J. Zucca Paul W. Kasameyer This paper was prepared for submittal to Geothermal Resources Council Annual Meeting Sparks, Nevada October 11-14, 1987 July 10, 1987 process disclosed, or represents that its use would not infringe privately owned rights. Refer- ence herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recom- , m5ndation. or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. DISCLAIMER This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency Thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. DISCLAIMER Portions of this document may be illegible in electronic image products. Images are produced from the best available original document. High Resolution Seismic Attenuation Tomography at Medicine Lnke Vol-, Calif0n-h * ~~ - UCRL--96157 John 1 Zucca and Paul W..Kasameyer DE87 013790 Lawrence Livermore National Laboratory, Livemore, Califomia SUMMARY Medicine Lake Volcano, a broad shield volcano about 5Okm east of Mount Shasta in northern California. produced rhylotic eruptions as recently as 400 years ago. Because of this recent activity it is of considerable interest to producers of geothermal energy. In a pint project sponsored by the Geothermal -...A - -e..--..____I. Research Program of the USGS and the Division of .&:- Geothermal and Hydropower Division of the US-DOE. the USGS and LLNL conducted an active seismic i.. experiment designed to explore the area beneath and arod the caldera. The experiment consisted of eight explosions detonated in a 50 Ian radius circle around the volcano recorded on a 11 x 15 km grid of 140 seismographs. The travel time data from the experiment have been inverted for structure and are presented elsewhere in this volume. In this paper we present the results of an inversion for 1/Q structure using t* data in a modified AM inversion scheme. Although the data are noisy, we find that in general attenuative zones correlate with low velocity zones. In particular, we observe a high 1/Q zone roughly in the center of the caldera at 4 km depth in between two large recent dacite flows. This zone could represent the still molten or partially molten source of the flows. Figure 1. Map showing location of Medicine Lake volcano Just east of the Cascades in northern California. The stars indicate the approximate locations of the eight INTRODUCTION explosions, which produced data recorded by 140 stations within the ellipse. The dotted lines indicate the For a number of years the USGS has used the refraction limes of Zucca, et. al., 1986. teleseismic travel-time residual method to study upper crustal magma bcdies. See their summary paper in this volume. For example, a recent study of the Mono The teleseismic method has two limitations: Craters area is descnied by Achaw, et al. (1986). In Teleseismic signals contain only low-frequencies, this method a tw+dimensional grid of seismic sensors limiting the resolution of the image to blocks 5-6 detects signals from a number of earthquakes. Only kilometers on a side. and the earthquakes may occur signals from distant events are studied, so that the only at a few azimuths, limiting the range of wavefront can be assumed planar when it enters the projections available. The USGS proposed to eliminate bottom of the region of interest. Under this these problems by using explosions placed uniformly assumption. the relative arrival time at each station is around the grid of seismometers. The explosions would the integral of the inverse of the seismic velocity along be placed far enough from the seismometers to produce the path through the cmst. For each earthquake a refracted energy coming upwards through the area of contour map of relative arrival times is produced, interest and close enough to preserve high frequencies which represents the projection of the velocity in order to enhance the resolution to blocks two structure along the angle of incidence of the wave. By kilometers on a side. We have extended this "high combining projections from a number of earthquakes at resolution" velocity technique to include a separate / different azimuths and distances, a three-dimensional inversion for attenuation. Instead travel times we \ image of the anomalous velocity is constructed. If a invert for variations in t*, which can be expressed large zone of partially molten material lies beneath the analogously to travel time as an integral of 1/Q along grid of stations, then it is seen on the image as a zone the raypath. Q is the quality and is thought of simply of anomalously low velocity. as the inverse of attenuation. R EXPERIMENTAL DESIGN SETTING FOR THE EXPERmENT LLNL was asked to participate in the project Medicine Lake volcano is a mio-Pliocene shield because of past involvement in seismic studies in the volcano located to the northeast of Mt. Shasta in the Medicine Lake area (Zucca et. al, 1986) and because the California Cascade Mountains (Figure 1). The original LLNL seismic system has complementary capabilities summit of the volcano collapsed and formed a 65 square to the USGS system. The USGS system consists of 120 kilometer caldera which has been buried by younger single component analog seismographs with independent flows originating from the caldera rim. Two of these recorders that are put in place and set to run for a flows are rhyolitic and may be as recent as 400 years specific time that the explosions are to occur. The old. Glass Mountain is the most extensive of these LLNL seismic system (Jarpe, et al. 1987) has flows. The chances are good that the magma chamber three-component sensors. The data are digit'ized at the that produced this flow is still active. The outlines of station and are transmitted to the central recording van the caldera and Glass Mountain are shown in Figure 2. via digital telemetry. The network can be started from the recording van, or set to trigger if an earthquake is Several lines of evidence suggested that Medicine detected. Lake Volcano was a good target for a detailed seismic imaging experiment. Teleseismic residual data (Evans, All 140 seismographs were deployed in a 15 km by 1982), gravity data (Finn and Williams, 1982), and 11 km grid with a 1.7 kilometer station spacing. The seismic refraction data (Zucca et al.. 1984) all show station distribution is shown in Figure 2. The 20 LLNL evidence for a shallow high velocity intrusion in the stations were used to cover the relatively inaccessible upper crust. The "Hot Spot," a small hot spring located areas around Glass Mountain, and the USGS stations near Glass Mountain, is the only surface expression of covered the remainder of Medicine Lake Volcano. the geothermal system which is presumed to be a result Eight 2,000 - 4,000 pound charges were detonated in a of an upper-crustal heat source (magma body) beneath single night in a circle with a 50 km radius about the' the Medicine Lake volcano. receivers, and more than 1100 seismograms were recorded to be used to produce the three-dimensional image of the caldera. In addition, the LLNL system was run in an event detecting mode for three weeks to determine if there is microseismicity or seismic noise A associated with the '%ot spot". None was detected. INTERPRETATION Two three-dimensional images of the crust under the caldera were produced using the method of Aki (1977). The USGS was responsible for producing the velocity image which is discussed below in comparison to the attenuation image. In a parallel effort, LLNL calculated estimates of the power spectra for each of the observed seismograms. Figure 3 shows an example of the seismograms. The picks labeled T1 are USGS picks of arrival time for the trough rather than the fit arrival. We estimated the spectrum of the arrival in a one second window located by using the USGS pick. By examining the slope of the power spectrum as a function of frequency, the relative attenuation along each travel path was estimated. (Figure 4). From the attenuation projections, the 3-D attenuation structure Figure 2. The location of the recording array is shown was also constructed. The details of these calculations on a schematic map of the Medicine Lake Volcano. The will be discussed in a paper being prepared by Evans and elliptical outline represents the caldera boundary, and Zucca. from east to west the irregular shapes show the outline of Class Mountain, Mt. Hoffman Glass flow, and Medicine Lake. These outlines also appear for reference in Figure 5 and 6. The scale is indicated by the caldera which is about 11 km wide. L' " I I 'I' '' 11 ' 8 ' Id' " 'AI a ~ ''4 Figure 3.